306K subscribers
4K photos
707 videos
17 files
4.58K links
Погружаемся в машинное обучение и Data Science

Показываем как запускать любые LLm на пальцах.

По всем вопросам - @haarrp

@itchannels_telegram -🔥best channels

Реестр РКН: clck.ru/3Fmqri
Download Telegram
📚 A practical approach to machine learning.

GitHub : https://github.com/GokuMohandas/practicalAI
👍1
A Gentle Introduction to Generative Adversarial Network Loss Functions

https://machinelearningmastery.com/generative-adversarial-network-loss-functions/
Adapt or Get Left Behind: Domain Adaptation through BERT Language Model Finetuning for Aspect-Target Sentiment Classification

https://arxiv.org/abs/1908.11860
1
Pytorch implementation of the paper "Class-Balanced Loss Based on Effective Number of Samples»

https://github.com/vandit15/Class-balanced-loss-pytorch

Class-Balanced Loss Based on Effective Number of Samples

https://github.com/richardaecn/class-balanced-loss
💬 Announcing Two New Natural Language Dialog Datasets

https://ai.googleblog.com/2019/09/announcing-two-new-natural-language.html

Coached Conversational Preference Elicitation

A dataset consisting of 502 dialogs with 12,000 annotated utterances between a user and an assistant discussing movie preferences in natural language.

https://ai.google/tools/datasets/coached-conversational-preference-elicitation

Accessing the Taskmaster-1 dataset

The full Taskmaster-1 dialog dataset has total 13,215 dialogs with 7708 written and 5507 spoken.
https://storage.googleapis.com/dialog-data-corpus/TASKMASTER-1-2019/landing_page.html
17th September In Moscow MegaFon office will host another meetup. Speakers from Mail.Ru, Altinity, Couchbase and MegaFon will talk about Statefull in Kubernetes. Free admission.
For details and registration : https://pao-megafon--org.timepad.ru/event/1056036/